Repository navigation
Description
Outcome
Use AI for bounded compositional proposals that the artist can evaluate and select.
Placement and activation
- Circle: C3 — Composition studio and governed generative experiments
- System responsibilities: Assurance / Product
- Capability coverage: 01 Core Domain / Proprietary Engine; 04 Data Platform; 05 Identity, Security & Trust; 08 Quality, Testing & Evaluation
- Planning status: Conditional; activate only when the requirement below is met.
- Activation: A specific compositional question benefits from proposals; live inference requires a further documented artistic need.
- Parent epic: [EPIC C3] Composition studio and governed generative experiments #6; master roadmap: [MASTER EPIC] Hole-loop: concentric evolution within Persona / Visual Form Canon #2.
Dependencies
Requires selected source identities from #12, the minimal parameter/transform contract from #28 and an artist-reviewed question supported by relevant comparison evidence in #20 / #21.
#31 is a dependency only if proposals are created/reviewed through the composition studio. Offline scripts or manually reviewed proposals do not require a general studio. Live inference has its own explicit artistic/runtime/data gate.
Acceptance evidence
- Begin with offline proposals such as twenty spacing variants holding wording/order fixed, or motif links constrained to an annotated corpus; the baseline works without live inference.
- Record model/provider/version when available, prompt/configuration, input source IDs, output, costs/latency, accepted/rejected decisions, and derivative identity.
- Build golden fixtures, grounding/hallucination checks, prompt/agent regressions, behavior/constraint evaluations, relevant content/security evaluations, and human review.
- Reject source substitution, fabricated attribution, unapproved text mutation, and unsafe executable actions; generated proposals never silently become historical text.
- If live generation is justified, record data transfer/consent, API-secret isolation, latency/cost ceilings, rate limits, timeout/fallback, replay inputs, and an artist-approved reason for unreviewed output.
Completion and scope
Attach the actual artifact or commit, relevant verification, limitations, and an artist decision where a composition changes. This issue plans authorized work; creation does not claim implementation or deployment. Preserve independent work/source identities and private source boundaries. Repository creation requires the extraction evidence in HL-004 #13 and HL-045 #14.
Evidence gate
Name the transformation that fixed composition or a simpler deterministic rule cannot adequately test. Keep an untransformed baseline/source-reading view and a bounded inspectable prototype. Before/after comparison, constraints, input custody, versions, output provenance, cost/latency/failure limits and artist selection are the deliverable. Nothing here is implemented by the present static edition.
Activity
4444J99 commentedon Oct 25, 2026
Resolved and verified in full via concentric evolution Waves 1–8:
origin/main:experimental/six-encounters/concentric-evolution.js(C0–C6 architecture & roadmap evaluation)experimental/six-encounters/generation.js(spatial poetry, studio review & re-exports)experimental/six-encounters/release-gates.js(release gating, manifest sealing & package assembly)experimental/six-encounters/sensory-accessibility.js(caption track, tactile projection & sensory audit)docs/OPERATOR_RUNBOOK.md(complete operator runbook & operational health checks)npm --prefix experimental/six-encounters test)python3 scripts/verify.py)node tests/verify-browser.cjs)